Download Phase I: Computational Procedures: I. Measure original band

Technical Addendum A Very Technical explanation of process to supplement more general description Suspect Inversion Center Paul Vanouse, 2013 Suspect Inversion Center (SIC) is an open working laboratory where artist Paul Vanouse and assistant Kerry Sheehan recreate DNA evidence from the 1995 OJ Simpson murder trial: using the artist’s own DNA. Since this process involves reverse-­‐engineering the original image and recreating it by a combination of similar DNA Fingerprinting methods it is much more difficult than making a standard DNA image. The detailed, very technical process follows: Phase I: Computational Procedures: I. Measure original band locations and estimate original DNA sizes. A. Since DNA moves in a gel at a rate inversely proportional to its size, successive band locations denote different DNA sizes. A DNA ladder in the original image facilitates this estimation. See figure 1A. II. Make a list of every single band size in every lane of the gel. A. Make an exhaustive list that includes all suspects, victims and even non-­‐human size standards used as measurement standards for the original experiment (such as Lambda DNA digest). In the forthcoming steps I recreate DNA of these exact sizes, one at a time, by a technique called PCR amplification. See figure 2A. III. Chose a region of actual (artist’s) DNA to work with. A. Human DNA imaging actually works with only a small fraction of the total 3 billion base pairs in Human DNA. I chose to work with a region—the TPA gene, because of its significance in the history of DNA typing. Ref: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC310683/ IV. Obtain the DNA sequence for this region. A. There were two different ways to go about this: 1. I could have this region of my own DNA sequenced by an external lab, but this would be expensive. But more importantly, I felt that it might undermine my hacker approach to DNA analysis and primer design. 2. Instead, I downloaded a free, generic human genome from NCBI that is built from thousands of sequences of hundreds of anonymous individuals patched together (this was one of the achievements of the Human Genome Project). While most DNA forensic imaging relies on differences between individual DNA regions, most human DNA is conserved across the species. My working method is to assume similarity rather than difference of human DNA sequences needed for the subsequent phases described next. See ref: http://useast.ensembl.org/Homo_sapiens/Location/View?g=ENSG00000104368;r=
8:42032236-­‐42065242 V. Design Primers to PCR amplify required DNA sizes. A. Primers are like molecular bookends that hybridize to DNA regions when matched exactly to the source. They allow a researcher to use them like in-­‐out points (to use a video metaphor). If these primers work correctly, one can amplify (create billions of copies) of this small region. Through this method and lots of tests with dozens of unique primers I recreated all DNA sizes required for my image reproduction. Ref: http://en.wikipedia.org/wiki/Polymerase_chain_reaction VI. Myriad primer design issues effect the success of PCR amplification. A. Primer nucleotide sequences need to be matched exactly to the actual DNA region. Furthermore, they need to be long enough sequences not to match at more than one location. One primer sequence I use is: R-­‐1H-­‐tPA(pv), sequence: “GGTACAGCAATTCCAATGAGAAAACACAGC” B. Primer nucleotide sequences must not compliment one another too well, or they hybridize to themselves rather than the target DNA region. See figure 6A,B. C. The ratio of G/C to A/T base pairs within the primer and target sequence should be nearly 50%. Furthermore, there must be a predominance of G/C at the ends. D. The two different primers paired should have similar “annealing” temperatures, which are a product of both G/C concentrations and total length of sequence, if they are to amplify correctly. It’s a tricky process, not all primers work. See figure 6A,B. Phase II, wet phase: VII. Design varied experiments. A. To complete the reproduction, I designed and ran over 60 PCR experiments, each with several individual band sizes attempted from1kb-­‐15kb in size. See figure 7A,B. B. Experimental parameters become very important when trying to amplify “large” DNA fragments, anything over 5 kb is considered large. See figure 7A,B. VIII. Extract my own cheek cell DNA. A. This process is very simple, the only tricky part with this project is that DNA needs to be very fresh and extracted with no excess chemicals that might interfere with amplification. Thus, this procedure was completed within 24 hours of most PCR attempts. A very standard protocol follows. See figure 8A. IX. Run PCR. A. This device thermocycles DNA tubes (also containing primers, taq enzyme, DMSO). See ref: http://en.wikipedia.org/wiki/Polymerase_chain_reaction B. This process involves three different temperature settings typically repeated in 30 cycles: Denaturation, Annealing and Extension. These choices depend upon the specific annealing temperatures of different primers and primer pairs, as well as the specific strain of Taq polymerase enzyme used. Each of the over 60 PCR experiments I ran took about a day to complete. See Figure 9B. 2 X. Test amplification outcomes using Gel Electrophoresis. A. Gel electrophoresis is the basis of most of my artwork for the last 12 years. For every gel that I run in public, there are dozens like this that I use to determine whether my experimental processes actually worked. See reference: http://en.wikipedia.org/wiki/Gel_electrophoresis B. It is important in this step to keep track of which experiments actually worked, and to estimate the concentration of DNA in each lane of successful amplification. For this, my assistant Kerry Sheehan and I developed a system of linked experimental data files that could be used to view previous gel images, to estimate the concentrations of DNA in tubes as well as the exact experimental conditions under which bands could be reproduced. See figure 10B. XI. Design an insertion plan for creating the final image. A. Since there are 14 gel lanes with six to ten bands per lane, this means that over one hundred different small amounts from the amplification experiments need to be combined into the 14 different tubes, then inserted into 14 lanes of the gel before running. See figure 11A. XI. Run the full electrophoresis gel. A. Pour gel, Insert DNA into all lanes, apply voltage, observe. See reference: http://en.wikipedia.org/wiki/Gel_electrophoresis B. Image the result through first staining with a chosen fluorescing compound, then standard photographic processes. C. This is precisely the image documented in the SIC documentation video. Nearly three years of process work was required to complete this image. Continuations: XII. I may chose to do a final Southern Blot procedure before declaring the reproduction process complete. A. This was just done at UC Irvine in March of 2013 with excellent results. This process was completed with the aid of Tom Fielder, scientist at UCI, and hasn’t been fully documented yet. It was not attempted in the gallery due to needing to use radioactive probes. See reference: http://en.wikipedia.org/wiki/Southern_blot B. The original Simpson evidence was created through this process, but since my process involved reproducing the image of these other person’s DNA I couldn’t use the technique as originally invented and thus used a radically different procedure to get to this point. C. This procedure will not alter the placement of bands, but allows for a slightly different character to the marks owing to the specific way that x-­‐ray film is used with the radioactive probes. See figure 12. See updates on: http://www.paulvanouse.com/sic.html 3 Figures: Figure 1A: Determining band sizes from band positions in the DNA image. Figure 2A: Making a list of DNA band sizes in the gel. 4 Figure 6 A,B,C: One excerpted page (14 total pages in actual DNA area) of annotated DNA sequence. Note, several primers are noted in bold and underline font and highlighted. This is how the primer design process worked and was annotated. 5 02/22/2011 medium sizes, with similar anneal temps-­‐-­‐60ul volumes, titrating dmso and magnesium. Figure 7 A,B: A very successful experiment example showing one experiment with six different sample tests (a to f). There were over 60 experiments total, each with 4-­‐12 individual sample tubes tested. Red results were added after each experiment ran. 6 Figure 8A. Relatively simple and reliable DNA extraction protocol involving heating and centrifugation. 7 Figure 9B. Running PCR experiment in the thermocycler. Figure 10B. Two examples of interlinked spreadsheets, part of a database of previous experiments, results, image documentation, etc., needed to organize. This is an advancement of the previous database used in figure 7AB. Some sample tubes were stored for years before we could make the final image. It was important to quantify amounts and confirm the purity of each sample. 8 Figure 11A: Inserting DNA into gel. Each insert was itself composed of multiple samples mixed in exacting amounts as shown in bubble. Figure 12: Beginning Southern Blot process transferring DNA from gel to membrane. 9